Creating a function that uses contains to check a string then convert to syms

The title wording is a bit confusing but here is the rundown.

I have an input that looks like this:

inputmsg = 'Enter a function of x and y to be approximated:\n';
f = input(inputmsg,'s'); % Take input as string
    variable = ["x","y"]; % Patterns to check for
    existXY = contains(f,variable) % If the input contains an 'x' or 'y', it will have logical 1
    
    % If f doesn't contain 'x' or 'y', repeat
    while ( existXY == 0 ) 
        fprintf('\nPLEASE ENTER A FUNCTION CONTAINING X AND Y\n');
        f = input(inputmsg,'s');
        variable = ["x","y"];
        existXY = contains(f,variable)
    end
f = str2sym(f); % Convert f back to symbol
% do maths

This works, but looks very clunky if put half way through a script.

A while back, I was shown how to check for numbers similar to this.

When trying to create a function, if I do either one of these, I get the error thats in the comments.

f = inputFunc('Enter a function of x and y to be approximated:\n','s');

function inputFunc(n)
% TOO MANY INPUT ARGUMENTS

f = inputFunc('Enter a function of x and y to be approximated:\n');

function inputFunc(n)
% TOO MANY OUTPUT ARGUMENTS

How would I lay out the function header to properly take in the arguments?

[dpb]

...    
variable = ["x","y"]; % Patterns to check for
existXY = contains(f,variable) % If the input contains an 'x' or 'y', it will have logical 1
while ~existXY      % existXY is a logical variable, treat it as such...
  fprintf('\nPLEASE ENTER A FUNCTION CONTAINING X AND Y\n');
  f = input(inputmsg,'s');
  %variable = ["x","y"];        % variable is never changed, no need to duplicate
  existXY = contains(f,variable);
end

but, really don't need variable at all as a separate variable here and your input isn't forgiving at all if user uses X,Y instead of x,y...

...    
existXY = contains(lower(f),variable)                       % allow either case
while ~existXY
  fprintf('\nPLEASE ENTER A FUNCTION CONTAINING X AND Y\n');
  f = input(inputmsg,'s');
  existXY = contains(lower(f),["x","y"]);
end

A potential issue I see is the above is T for either x or y; and your code doesn't distinguish between which or both are in the expression. Hence, seems will be difficult to know what to do with it.

You don't give sufficient detail on just what you then are trying to do to know exactly what is needed...and I don't have Symbolic TB so I "know nuthink!" about it when get to that point...

But, for function headers to accept variable numbers of inputs, outputs, look at varargin, varargout

[Richard Cheung]

Adding the second variable is my bad.

I have defined lowercase x and y as symbols so upper case wouldn't work with the rest of the code. If capitals are entered, the logical is 0, so it will ask again. I'd have to edit my warning message so that lowercases appear.

syms x y h k

Using lower(f) will let X and Y pass, but the rest of the code won't accept those variables.

I was playing around and found that if I input something like "x + b", it would still pass, so I'll need to look into that.

I guess I should probably ask a new question about the symbolic toolbox before continuing with this approach since without knowing whether theres a way to treat special symbols like cos etc as its own thing, it would be hard to get this to properly work.

[dpb]

Well, for your users' sake then at least take the input and cast to lower for them instead of making them reenter from scratch. Unless, of course, it's feasible to have something like

y=2x + A

We don't know what it is you're intending/trying to do so pretty difficult to guess what would need to do or whether the symbolic toolbox would necessarily even be needed.

[Richard Cheung]

The most I can say is that it's necessary for my programming project.

Making a function like this is probably a bit overkill since all I want to do is return an error if a bad input is entered, which matlab will already do, albeit it'll look a bit ugly.